局部保持鉴别投影及其在人脸识别中的应用.pdfVIP

第35 卷第2 期 电 子 与 信 息 学 报 Vol.35 No.2 2013 年2 月 Journal of Electronics Information Technology Feb. 2013 局部保持鉴别投影及其在人脸识别中的应用 赵振华* 郝晓弘 (兰州理工大学电信学院 兰州 730050) (甘肃工业先进过程控制重点实验室 兰州 730050) 摘 要：针对流形学习在人脸识别中的应用，该文提出基于局部保持投影(Locality Preserving Projection, LPP) 的监督线性维数约简方法。利用样本的类别信息，将LPP 的最近邻图分解为类内图和类外图，通过优化，最优保 持同类数据固有的局部邻域关系，缩小数据之间的距离，同时最大化不同类数据之间的距离，从而增大各类数据分 布之间的间隔，提高了嵌入空间的辨别能力。此外，在构建图的过程中采用了自适应邻域，增强了对数据分布稀疏 性的表征。在Extended Yale B 和CMU PIE 两个开放人脸数据库上进行了试验，验证了算法的有效性。 关键词：人脸识别；流形学习；自适应邻域；监督学习；局部保持投影 中图分类号：TP391.4 文献标识码： A 文章编号：1009-5896(2013)02-0463-05 DOI: 10.3724/SP.J.1146.2012.00601 Linear Locality Preserving and Discriminating Projection for Face Recognition Zhao Zhen-hua Hao Xiao-hong ( College of Electrical Engineering and Information Engineering, Lanzhou University of Technology, Lanzhou 730050, China) (Key Laboratory of Gansu Advanced Control for Industrial Processes, Lanzhou 730050, China) Abstract: A novel supervised linear method based on Locality Preserving Projection (LPP) of reducing dimensionality is proposed for face recognition. In this study, the nearest neighbor graph of LPP is split into within-class graph and between-class graph according to the class label information of samples. After optimizing, the intrinsic local neighbor structure of the samples of same class is maintained and the distances between them are decreased. Meanwhile, the distances between the samples of different class are maximized to increase the space of the distribution of all kinds of samples, and thus the discriminabili